The Yammer home feed drives discovery and engagement across communities in a network. The feed relies on algorithms to determine which posts are the most relevant to you. Every person’s home feed is different, because we all care about different things. The Yammer home feed algorithm scans all Yammer messages that are available to you every time you visit the feed, personalizes the experience and helps you find the conversations that are relevant to you.
And depending on how active and the size of your network, the number of conversations may be overwhelming, especially to new members. So, how does Yammer help you find the posts that are particularly relevant to you? What impacts the feed and how can you help influence what is delivered? Let’s take a closer look...
Yammer looks at who you follow and what Yammer communities you’ve joined. Yammer also looks at the conversations happening across your entire network. The more signals you provide to Yammer, the better job we can do at recommending content you care about.
People and communities. Whenever you visit the Yammer home feed, we get a list of all the of people you engage with most frequently. These are the people that you interact with in Yammer, Outlook, Microsoft Teams, or other Microsoft 365 tools.
We also pull the list of communities you participate in. For example, if you visit a community, or reply to a post from that community, then we increase the "affinity score" between you and that community.
Collectively, we use a list of the closest people and the top communities and pull all the conversations from those two groups. We examine each conversation and pass them through our machine learning model. For example, communities with more participation receive a different weight for the algorithm. We also look at information regarding the user who posted the conversation. For example, how many followers does this user have, how much interaction you have with that person, or how active they are in the network.
Conversation attributes. Attributes such as who the author is, how long ago the message was posted, who replied, who reacted and how many people have reacted to the message are all examples of the ways we evaluate every conversation to determine its relevancy to you. Other contributing factors include whether your post includes pictures and videos or whether it is a poll or a question.
All of the posts are ranked based on relevancy and the likelihood that you’ll like or reply to them. Sometimes, these recommendations will make absolute sense to you because it’s a message from your boss, or your closest coworker.
We add all these attributes into the machine learning ‘blender’, and return a prioritized list of conversations for your feed. Additionally, we will help you discover unseen content that is relevant. We think of it as expanding your peripheral vision. Eventually, you’ll begin to see posts from other communities and coworkers that derive from this prioritized list of relevant conversations.
Once we rank all threads from the communities you belong to and the people you follow, we remove any threads from communities that you have muted. If your network admin has chosen to feature a conversation, we will show you that “Featured Conversation” near the top of your feed. This is how we assemble the final feed that you see on your Yammer Home page.
How we personalize your home feed
How do featured conversations rank in the feed? A featured conversation is a post that the community admin has boosted for a specific time window. If there is more than one featured conversation available to you, we will show the soonest-to-expire one.
What about SharePoint News articles? We also include SharePoint News articles if there is one available to you. You can expect less than 1% of the feed with SharePoint news article.
What else could show up in my feed? Recommendations for you to discover more people to follow or communities to join. We only show this module if we predict there is a high likelihood of you accepting our recommendations.
Yammer will also show you ranked trending content. This may be content you are not subscribed to, but Yammer thinks it could be interesting based on the high number of reactions a conversation is getting from many people across the org.
The Yammer home feed relies on signals (previous actions you have taken) to understand what you find interesting. These are the actions you can take to influence the content in your feed.
We are constantly learning from you about how you use the feed. You can expect our feed relevancy to continue to improve over time.
Stay tuned for more details as we continue fine tuning the Yammer feed for you.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.